, and making dynamically migratable application programs easy to write. This paper describes object arrays, a construct which makes dynamically migratable applications easier to write, and a simple strategy for migrating load on a workstation cluster.

We investigate the potential of workstation clusters for use in high performance computation for some selected applications. Currently, the network speed found in most of the existing systems is quite low, but higher speed networks are already emerging in the market. We present four parallel

the SCI technology and demonstrates the use of one comercially solution for SPARC workstation clusters. Detailed performance information is included. 1 Introduction SCI technology is one of the research directions of the computer architecture group at the University of Technology in Chemnitz. Recently

be identified in advance. We explore a technique called dynamic coscheduling (DCS) which produces emergent coscheduling of the processes constituting a parallel job. Experiments are performed in a workstation environment with high performance networks and autonomous timesharing schedulers for each CPU

The goal of this paper is to determine how efficiently we can implement an adequate parallel programming environment on a workstation cluster without modifying the existing operating system. We have implemented a runtime environment for parallel programs and gang-scheduling on a workstation cluster

Advances in network and processor technology have greatly changed the communication and computational power of local-area workstation clusters. However, operating systems still treat workstation clusters as a collection of loosely-connected processors, where each workstation acts as an autonomous

This paper investigates the use of workstation clusters as parallel processors. We concentrate on the development of parallel versions of existing sequential programs to exploit this parallel processing capability of clusters of workstations

We present a system for functional parallel computing on distributed memory machines with dynamic load balancing and recovery. Every node can perform any computation and is client and server at the same time. The nodes themself are multithreaded to overlap communication with computation and minimize latency. Furthermore, the calling convention is similar to threads on shared memory machines to allow a uniform usage of procedure calls on different architectures. The communication and computation is strictly separated for easy maintenance of the user program. Finally, we show an example and some timing results. 1 Introduction The power of modern computers has increased a lot during the last years. This is due to new concepts (like RISC) and better technologies resulting in higher clock rates. However, the current technology has reached a point where further improvements are getting harder. A promising concept for further improvements is parallelism. It allows the simultaneous use of mor...

We present novel hard- and software that efficiently implements communication primitives for parallel execution on workstation clusters. We provide low communication latencies, minimal protocol, zero operating system overhead, and high throughput. With this technology, it is possible to build

Despite the emergence of high speed LANs, the communication performance available to applications on workstation clusters still falls short of that available on MPPs. A new generation of efficient messaging layers is needed to take advantage of the hardware performance and to deliver